{"id":"W4415460301","doi":"10.3390/higheredu4040065","title":"Harnessing Large Language Models for Scalable and Effective Formative Assessment in Higher Education: A Review","year":2025,"lang":"en","type":"article","venue":"Trends in Higher Education","topic":"Higher Education Learning Practices","field":"Social Sciences","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"The Scarborough Hospital; University of Toronto","funders":"Rhodes Scholarships","keywords":"Formative assessment; Context (archaeology); Transformative learning; Scalability; Higher education; Implementation","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001426769,0.000162358,0.0002712163,0.0004433199,0.0002918023,0.0001499293,0.0001812892,0.0001209341,0.001210255],"category_scores_gemma":[0.0001058137,0.0001699937,0.00004458048,0.001508561,0.0000770134,0.001246785,0.00003442841,0.000285883,0.000008582182],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007586718,"about_ca_system_score_gemma":0.001361623,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001134491,"about_ca_topic_score_gemma":0.0004235509,"domain_scores_codex":[0.998113,0.0006134494,0.0003660526,0.000367856,0.0002118988,0.000327689],"domain_scores_gemma":[0.9987648,0.0005254863,0.0002386361,0.0002243574,0.0001705907,0.00007613275],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00004093726,0.002418153,0.04164404,0.001502395,0.00004968856,4.257648e-7,0.0756219,0.00005815069,0.00002406369,0.5625674,0.05101829,0.2650546],"study_design_scores_gemma":[0.000619938,0.00002898284,0.3209091,0.002403351,0.00007665688,3.448679e-7,0.00847591,0.00007637198,0.000009154279,0.007473131,0.6596438,0.0002833353],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.1023711,0.05668105,0.0002692057,0.218444,0.01015392,0.003401815,0.00001867457,0.0002113005,0.6084489],"genre_scores_gemma":[0.8665547,0.0005325151,0.00223691,0.002246457,0.0002978611,0.001903008,0.00007961579,0.00001554643,0.1261334],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7641836,"threshold_uncertainty_score":0.9997028,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04975622032729901,"score_gpt":0.4643202307419215,"score_spread":0.4145640104146225,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}